Volt Normals · VN-1
Climatology that admits the trend
The documented, persistent mispricing in weather markets is stale climatology: 10- and 30-year normals lag the warming trend — and the warm tails are fattening faster than the means. Volt Normals fit an exact quantile regression on year per station, per day-of-year, per quantile, and evaluate it at this year. Free to cite with attribution ("Volt Normals (Voltcast)" + link).
Today — trend-adjusted vs stale, °C
Full-year CSV, all stations →| Station | Tmax normal (VN-1) | Stale climatology | Stale bias | Q90 today | Trend Q50 | Trend Q90 |
|---|---|---|---|---|---|---|
| EBBR Brussels Airport | 23.8 | 23.5 | +0.3 | 30.2 | +0.12°C/dec | +1.22°C/dec |
| EDDL Düsseldorf Flughafen | 24.0 | 24.9 | -0.9 | 31.0 | 0.00°C/dec | 0.00°C/dec |
| EDDM München Flughafen | 26.4 | 24.6 | +1.8 | 31.0 | +1.11°C/dec | 0.00°C/dec |
| EETN Tallinn Lennart Meri | 23.8 | 22.5 | +1.3 | 27.0 | +2.54°C/dec | 0.00°C/dec |
| EFHK Helsinki-Vantaa | 25.0 | 22.9 | +2.1 | 28.7 | +2.33°C/dec | +0.20°C/dec |
| EGLC London City Airport | 23.0 | 22.7 | +0.3 | 29.2 | 0.00°C/dec | +0.83°C/dec |
| EGLL London Heathrow | 23.5 | 23.9 | -0.4 | 30.3 | -0.73°C/dec | +0.33°C/dec |
| EHAM Amsterdam Schiphol | 22.7 | 22.3 | +0.4 | 28.7 | +0.85°C/dec | +0.50°C/dec |
| EIDW Dublin Airport | 19.3 | 19.4 | -0.1 | 23.8 | -0.43°C/dec | +0.25°C/dec |
| EKBI Billund Lufthavn | 22.4 | 21.7 | +0.7 | 28.5 | +1.20°C/dec | +1.18°C/dec |
| EKCH København Kastrup | 22.2 | 21.8 | +0.4 | 26.3 | +1.38°C/dec | 0.00°C/dec |
| ENBR Bergen Flesland | 18.4 | 17.8 | +0.6 | 25.9 | +1.25°C/dec | +1.75°C/dec |
| ENGM Oslo Gardermoen | 21.8 | 21.5 | +0.3 | 29.0 | +1.00°C/dec | +1.60°C/dec |
| ENTC Tromsø Langnes | 18.7 | 15.5 | +3.2 | 23.3 | +4.30°C/dec | +3.09°C/dec |
| ENVA Trondheim Værnes | 22.0 | 19.3 | +2.7 | 31.8 | +3.45°C/dec | +4.83°C/dec |
| ENZV Stavanger Sola | 18.5 | 18.4 | +0.1 | 25.1 | +0.50°C/dec | +0.33°C/dec |
| EPWA Warszawa-Okecie | 26.7 | 25.0 | +1.7 | 30.6 | +2.00°C/dec | +1.75°C/dec |
| ESMS Malmö Sturup | 22.1 | 21.6 | +0.5 | 27.0 | +1.00°C/dec | +0.42°C/dec |
| ESSA Stockholm-Arlanda | 21.8 | 22.6 | -0.8 | 27.1 | -0.82°C/dec | -1.40°C/dec |
| EVRA Rīga Airport | 22.4 | 21.6 | +0.8 | 26.8 | +0.42°C/dec | -0.14°C/dec |
| LBSF Sofia Airport | 31.6 | 29.1 | +2.5 | 36.3 | +2.00°C/dec | +1.54°C/dec |
| LDZA Zagreb Pleso | 29.5 | 27.0 | +2.5 | 33.0 | +1.21°C/dec | -1.43°C/dec |
| LEMD Madrid-Barajas | 36.6 | 34.4 | +2.2 | 41.0 | +1.46°C/dec | +3.12°C/dec |
| LFPB Paris-Le Bourget | 25.8 | 25.7 | +0.1 | 31.2 | +0.17°C/dec | +0.10°C/dec |
| LFPO Paris-Orly | 26.2 | 26.2 | 0.0 | 32.1 | 0.00°C/dec | +0.60°C/dec |
| LGAV Athens Eleftherios Venizelos | 33.8 | 32.9 | +0.9 | 38.3 | +0.80°C/dec | +1.82°C/dec |
| LHBP Budapest Liszt Ferenc | 31.0 | 28.3 | +2.7 | 35.2 | +2.50°C/dec | +2.14°C/dec |
| LIML Milano Linate | 32.4 | 30.6 | +1.8 | 34.8 | +2.00°C/dec | +0.86°C/dec |
| LIRF Roma Fiumicino | 31.7 | 29.4 | +2.3 | 34.5 | +2.17°C/dec | +2.33°C/dec |
| LJLJ Ljubljana Brnik | 29.0 | 28.5 | +0.5 | 31.9 | 0.00°C/dec | -1.56°C/dec |
| LKPR Praha-Ruzyně | 26.9 | 24.7 | +2.2 | 30.7 | +1.56°C/dec | +0.38°C/dec |
| LOWW Wien-Schwechat | 29.6 | 26.7 | +2.9 | 34.2 | +2.40°C/dec | +2.09°C/dec |
| LPPT Lisboa Humberto Delgado | 28.0 | 28.8 | -0.8 | 34.7 | 0.00°C/dec | +1.43°C/dec |
| LROP București Otopeni | 33.8 | 30.5 | +3.3 | 38.8 | +2.92°C/dec | +3.36°C/dec |
| LSZH Zürich Kloten | 27.7 | 25.5 | +2.2 | 32.1 | +1.67°C/dec | 0.00°C/dec |
| LTAC Ankara Esenboga | 31.2 | 29.9 | +1.3 | 35.0 | +0.90°C/dec | 0.00°C/dec |
| LTFM Istanbul Airport | 32.7 | 31.0 | +1.7 | 36.8 | +2.50°C/dec | +3.29°C/dec |
| LWSK Skopje Airport | 36.3 | 32.4 | +3.9 | 40.6 | +3.33°C/dec | +3.08°C/dec |
| LYBE Beograd Nikola Tesla | 31.5 | 29.4 | +2.1 | 36.4 | +2.50°C/dec | +0.93°C/dec |
| LYPG Podgorica Airport | 36.1 | 33.2 | +2.9 | 41.2 | +4.43°C/dec | +8.50°C/dec |
| LZIB Bratislava M. R. Štefánik | 29.7 | 27.7 | +2.0 | 33.1 | +1.71°C/dec | +0.75°C/dec |
| UUEE Moscow Sheremetyevo | 24.0 | 24.5 | -0.5 | 27.9 | 0.00°C/dec | -1.43°C/dec |
"Stale bias" = VN-1 minus the plain multi-decade mean for today's calendar day — the number a stale-normals model misprices by. An amber Q90 trend above the Q50 trend is the tail fattening faster than the mean.
Methodology — VN-1
- Window: the last 15 completed years of finalized daily Tmax/Tmin (ISD + METAR), ±7-day day-of-year window (15-day centred, leap-stable).
- Model: exact linear quantile regression (pinball loss, no regularization) of temperature on year — separately for q10/q50/q90 of Tmax and Tmin, so each quantile carries its own trend.
- Output: the fit evaluated at the current year, plus trend slopes in °C/decade. Rows carry sample sizes; method changes version (VN-2, …) and are never applied silently.
- Where it works: our degree-day settle projections use VN-1 beyond the ensemble horizon — the projection inherits the trend instead of a stale mean.
Use it
curl "https://voltcast.com/api/v1/normals" # today, all stations curl "https://voltcast.com/api/v1/normals/LFPB" # full-year curve, Paris-Le Bourget
Keyless and CORS-open. The station-exact forecast distributions, settle projections and degree-day trackers built on top live in Volt Temp (€99/mo) and on the Desk.